ggplot_lister: Function for creating sequential (or branching) ggplots

In cbedwards/cedwards: Collin Edwards' convenience functions

Function for creating sequential (or branching) ggplots

Description

For lectures or presentations, it can often be helpful to show a sequential series of plots. A common case: show the raw data, show the raw data and the fitted model, then show the raw data and the fitted model with an overlay of some sort of additional information. This process can become somewhat tedious for complex buildups, even with the relative streamlining ggplot provides. This function streamlines that process, and ggplot_list_saver further streamlines the workflow by also automatically saving the generated figures.

Usage

ggplot_lister(gglist, niceties = NULL)

Arguments

gglist	A list of ggplot layers, augmented with whole numbers to specify commands. First item in the list must be a ggplot() call. 0 specifies printing, negative integers specify removing the previous n layers before continuing. Each item MUST be either a ggplot() call, ggplot layer, 0, or negative whole number.
niceties	A single ggplot layer or a list of ggplot layers that should be applied to all plots, and are applied last. Common use case would be a list of axis labels, plot title, and theme. Defaults to NULL; in this case, no layers added. Note that this formulation is merely for convenience – these layers can also be specified in gglist, but would generally need to be added before the first 0 entry.

Details

The core functionality comes from the gglist argument. Layers are added sequentially; whenever a 0 is encountered in the list, ggplot_lister() saves the plot up to that point as a new entry in the output list. In some cases, it can be useful to backtrack ("Okay, but what if we fit the data with a smoothing spline instead of a linear regression?"). Rather than make separate calls to ggplot_lister(), you can backtrack by adding negative integers to the list. Adding negative value n removes the last n layers in the list before continuing.

For figure sets without backtracking, a simple workflow is to develop the final figure with all components before using ggplot_lister(). Then simply copy the ggplot() call and the entire collection of layers into a list for the gglist argument, replacing + with commas. Move any purely cosmetic layers (axis labels, themes, etc) to a separate list for niceties. Then, in the gglist argument, add new 0 entries to the list after each layer for which you want a new plot.

Value

List of ggplots, in order of creation.

Examples

library(tidyverse)
dat = data.frame(y = runif(30), x = runif(30))
out = ggplot_lister(gglist = list(ggplot(data = dat, aes(x = x, y = y)),
                                  geom_point(), 0,
                                  geom_path(),0, -1, geom_smooth(method = 'lm')),
                    niceties = list(xlab("turtle"), ylab("cow")))

out = ggplot_lister(gglist = list(ggplot(data = dat, aes(x = x, y = y)),
                                  geom_point(), 0,
                                  geom_path(),0,
                                  geom_point(aes(x = x+.2), col = 'blue')),
                    niceties = list(xlab("turtle"), ylab("cow")))

cbedwards/cedwards documentation built on Feb. 24, 2023, 6:53 p.m.